Bridging the Gap Between Detection and Tracking: A Unified Approach
The DOI matches its Crossref record; 28 of the 28 references it cites were found in a registry or an index.
Abstract
Object detection models have been a source of inspiration for many tracking-by-detection algorithms over the past decade. Recent deep trackers borrow designs or modules from the latest object detection methods, such as bounding box regression, RPN and ROI pooling, and can deliver impressive performance. In this paper, instead of redesigning a new tracking-by-detection algorithm, we aim to explore a general framework for building trackers directly upon almost any advanced object detector. To achieve this, three key gaps must be bridged: (1) Object detectors are class-specific, while trackers are class-agnostic. (2) Object detectors do not differentiate intra-class instances, while this is a critical capability of a tracker. (3) Temporal cues are important for stable long-term tracking while they are not considered in still-image detectors. To address the above issues, we first present a simple target-guidance module for guiding the detector to locate target-relevant objects. Then a meta-learner is adopted for the detector to fast learn and adapt a target-distractor classifier online. We further introduce an anchored updating strategy to alleviate the problem of overfitting. The framework is instantiated on SSD and FasterRCNN, the typical one- and two-stage detectors, respectively. Experiments on OTB, UAV123 and NfS have verified our framework and show that our trackers can benefit from deeper backbone networks, as opposed to many recent trackers.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
You Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803
Feature Pyramid Networks for Object Detection
Lin, T.-Y., Dollar, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.106 · Cited by 30,939
Learning to Compare: Relation Network for Few-Shot Learning
Sung, F., Yang, Y., Zhang, L., Xiang, T., Torr, P. H. S., & Hospedales, T. M. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00131 · Cited by 5,816
High-Speed Tracking with Kernelized Correlation Filters
Henriques, J. F., Caseiro, R., Martins, P., & Batista, J. (2015) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2014.2345390 · Cited by 5,753
Fully-Convolutional Siamese Networks for Object Tracking
Bertinetto, L., Valmadre, J., Henriques, J. F., Vedaldi, A., & Torr, P. H. S. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-48881-3_56 · Cited by 4,546 · cited 2×
“This suggests that our model can benefit from deeper network, which is opposed to many recent deep trackers [8, 3, 34] where introducing deeper network leads to similar or even worse performance.”
ResultOnline Object Tracking: A Benchmark
Wu, Y., Lim, J., & Yang, M.-H. (2013) · 2013 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2013.312 · Cited by 4,138
Visual object tracking using adaptive correlation filters
Bolme, D., Beveridge, J. R., Draper, B. A., & Lui, Y. M. (2010) · 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2010.5539960 · Cited by 3,406 · cited 2×
“, UMACE [41] and ASEF [6]) and later improved and adapted for tracking [5, 22, 23, 9, 8].”
BackgroundWu, Y., Lim, J., & Yang, M.-H. (2015) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2014.2388226 · Cited by 3,371
High Performance Visual Tracking with Siamese Region Proposal Network
Li, B., Yan, J., Wu, W., Zhu, Z., & Hu, X. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00935 · Cited by 3,023 · cited 8×
“UAV123 NfS SiamFC [3] 0.523 - SiamRPN [34] 0.571 - DaSiamRPN [59] 0.584 0.395 ECO [8] 0.537 0.470 Ours (SSD) 0.531 0.491 Ours (FRCNN) 0.586 0.515”
BackgroundMethodResultInfluentialDeep Learning for Generic Object Detection: A Survey
Liu, L., Ouyang, W., Wang, X., Fieguth, P., Chen, J., Liu, X., & Pietikäinen, M. (2020) · International Journal of Computer Vision · doi:10.1007/s11263-019-01247-4 · Cited by 2,965 · cited 2×
“Object detection is the task of locating and classifying objects in an image to a set of predefined classes [11, 13, 40, 15, 39].”
BackgroundDeep Learning for Generic Object Detection: A Survey
Liu, L., Ouyang, W., Wang, X., Fieguth, P., Chen, J., Liu, X., & Pietikäinen, M. (2018) · arXiv · doi:10.48550/arxiv.1809.02165 · Cited by 2,965
Learning Multi-domain Convolutional Neural Networks for Visual Tracking
Nam, H., & Han, B. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.465 · Cited by 2,806 · cited 2×
“Modules and designs in object detection algorithms are adopted to either improve accuracy (e.g., bounding box regression used in [21, 45, 34]) and/or enhance stability (e.g., re-detection mechanism used in [27]) of tracking.”
MethodECO: Efficient Convolution Operators for Tracking
Danelljan, M., Bhat, G., Khan, F. S., & Felsberg, M. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.733 · Cited by 2,805 · cited 5×
“UAV123 NfS SiamFC [3] 0.523 - SiamRPN [34] 0.571 - DaSiamRPN [59] 0.584 0.395 ECO [8] 0.537 0.470 Ours (SSD) 0.531 0.491 Ours (FRCNN) 0.586 0.515”
BackgroundMethodResultInfluentialExploiting the Circulant Structure of Tracking-by-Detection with Kernels
Henriques, J. F., Caseiro, R., Martins, P., & Batista, J. (2012) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-33765-9_50 · Cited by 2,310 · cited 2×
“For example, the correlation filters are previously applied in object detection (e.g., UMACE [41] and ASEF [6]) and later improved and adapted for tracking [5, 22, 23, 9, 8].”
BackgroundMethodBeyond Correlation Filters: Learning Continuous Convolution Operators for Visual Tracking
Danelljan, M., Robinson, A., Shahbaz Khan, F., & Felsberg, M. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46454-1_29 · Cited by 1,853
GOT-10k: A Large High-Diversity Benchmark for Generic Object Tracking in the Wild
Huang, L., Zhao, X., & Huang, K. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2957464 · Cited by 1,809 · cited 3×
“However, the task is still challenging due to several factors such as object interac-∗ tions, cluttered background, occlusions and target deformation [43, 24].”
BackgroundMethodSelf-citationLaSOT: A High-Quality Benchmark for Large-Scale Single Object Tracking
Fan, H., Lin, L., Yang, F., Chu, P., Deng, G., Yu, S., Bai, H., Xu, Y., Liao, C., & Ling, H. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00552 · Cited by 1,742 · cited 1×
“It is currently an active research area of computer vision and significant progresses have been made in recent years [57, 31, 30, 12].”
BackgroundATOM: Accurate Tracking by Overlap Maximization
Danelljan, M., Bhat, G., Khan, F. S., & Felsberg, M. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00479 · Cited by 1,591 · cited 2×
“Similar examples include the bounding box regression [13, 49] used in MDNet [45, 26] and GOTURN [21], structured output SVM [4] applied in Struck [17], region proposal networks [15] used in SiamRPN [34, 59, 32] and the precise ROI pooling [25] used in the ATOM tracker [7], to name a few.”
MethodTrackingNet: A Large-Scale Dataset and Benchmark for Object Tracking in the Wild
Müller, M., Bibi, A., Giancola, S., Alsubaihi, S., & Ghanem, B. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01246-5_19 · Cited by 1,192
Acquisition of Localization Confidence for Accurate Object Detection
Jiang, B., Luo, R., Mao, J., Xiao, T., & Jiang, Y. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01264-9_48 · Cited by 1,059
Antoniou, A., Edwards, H., & Storkey, A. (2018) · arXiv · doi:10.48550/arxiv.1810.09502 · Cited by 690
Wang, Q., Teng, Z., Xing, J., Gao, J., Hu, W., & Maybank, S. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00510 · Cited by 604 · cited 1×
“OTB-2015 SiamFC [3] 0.607 0.582 CFNet [53] 0.611 0.568 DSiam [16] 0.656 - RASNet [54] 0.670 0.642 SiamRPN [34] - 0.637 DaSiamRPN [59] - 0.658 SA-Siam [18] 0.677 0.657 ECO [8] 0.709 0.694 MemTrack [58] 0.642 0.626 Ours (SSD) 0.637 0.620 Ours (FRCNN) 0.656 0.647 Table 4: State-of-the-art comparison…”
InfluentialLearning Dynamic Memory Networks for Object Tracking
Yang, T., & Chan, A. B. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01240-3_10 · Cited by 382
Learning to Localize Objects with Structured Output Regression
Blaschko, M. B., & Lampert, C. H. (2008) · Lecture Notes in Computer Science · doi:10.1007/978-3-540-88682-2_2 · Cited by 324 · cited 3×
“Similar examples include the bounding box regression [13, 49] used in MDNet [45, 26] and GOTURN [21], structured output SVM [4] applied in Struck [17], region proposal networks [15] used in SiamRPN [34, 59, 32] and the precise ROI pooling [25] used in the ATOM tracker [7], to name a few.”
BackgroundMethodJung, I., Son, J., Baek, M., & Han, B. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01225-0_6 · Cited by 297
End-to-End Representation Learning for Correlation Filter Based Tracking
Valmadre, J., Bertinetto, L., Henriques, J., Vedaldi, A., & Torr, P. H. S. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.531 · Cited by 274
Cited by
Deep Learning for Visual Tracking: A Comprehensive Survey
Marvasti-Zadeh, S. M., Cheng, L., Ghanei-Yakhdan, H., & Kasaei, S. (2022) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2020.3046478 · Cited by 405
GlobalTrack: A Simple and Strong Baseline for Long-Term Tracking
Huang, L., Zhao, X., & Huang, K. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i07.6758 · Cited by 279
Self-citationTracking by Instance Detection: A Meta-Learning Approach
Wang, G., Luo, C., Sun, X., Xiong, Z., & Zeng, W. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00632 · Cited by 206
Siam R-CNN: Visual Tracking by Re-Detection
Voigtlaender, P., Luiten, J., Torr, P. H. S., & Leibe, B. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00661 · Cited by 115
Metadata
- Authors
- Huang, L., Zhao, X., & Huang, K.
- Venue
- 2019 IEEE/CVF International Conference on Computer Vision (ICCV)
- Publisher
- IEEE
- Published
- 2019-10
- Pages
- 3998-4008
- OpenAlex
- W2986235162
- Cited by
- 52
- References
- 28
- FWCI
- 3.04
- Type
- conference paper
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.